Re: [ws-rx] NEW ISSUE: What is the effective policy when RM policyassertion is present at multiple places?

[Anish Karmarkar <Anish.Karmarkar@oracle.com>]

One comment inlined below.

-Anish
--


Christopher B Ferris wrote:
> 
> Anish,
> 
> What the resolution to i021 said was that for the ***extensibility 
> elements*** of the RMAssertion that those
> present at the port trumped the extensions at the binding, ad nauseum. 
> It said nothing about the
> meaning of wsp:Optional which is consistent IMO with WS-Policy in that 
> it is shorthand for a policy
> statement that contains an ExactlyOne containing two Alls, one with, and 
> one without the RMAssertion.
> 
> e.g.
> 
> <wsp:Policy>
>   <wsrm:RMAssertion wsp:Optional="true"/>
> </wsp:Policy>
> 
> is equivalent to:
> 
> <wsp:Policy>
>   <wsp:ExactlyOne>
>     <wsp:All>
>       <wsrm:RMAssertion/>
>     </wsp:All>
>     <wsp:All>
>     </wsp:All>
>   </wsp:ExactlyOne>
> </wsp:Policy>
> 
> When you attach policy to a port, it applies to that port. It means that 
> RM is optional for all messages
> sent to that port. When you assert a policy statement at the message 
> level, it asserts for that message
> level. You can't undo it at the port level.
> 

Ok, so you are saying that my 'assumption/impression which follows from 
the current text' is wrong. I was hoping for that.
If that is so, then does WSP say that somewhere. If so, can u pl. point 
me to it. Thx!

> IMO, there is no issue here. I propose we close with no action.
> 

<snip/>